An efficient framework for automated screening of Clinically Significant Macular Edema

The present study proposes a new approach to automated screening of Clinically Significant Macular Edema (CSME) and addresses two major challenges associated with such screenings, i.e., exudate segmentation imbalanced datasets. proposed replaces the conventional based feature extraction by combining pre-trained deep neural network meta-heuristic selection. A space over-sampling technique is being used overcome effects skewed datasets accomplished k-NN classifier. role each data-processing step (e.g., class balancing, selection) limiting region interest fovea on classification performance are critically analyzed. Finally, selection implication operating points Receiver Operating Characteristic curve discussed. results this convincingly demonstrate that following these fundamental practices machine learning, basic classifier could effectively accomplish CSME screening.